Susceptibility of Candida Glabrata Biofilms to Echinocandins: Alterations in the Matrix Composition

BIOFOULING, 2018 VOL. 34, NO. 5, 569–578 https://doi.org/10.1080/08927014.2018.1472244

Susceptibility of Candida glabrata biofilms to echinocandins: alterations in the matrix composition

Célia F. Rodrigues , Maria Elisa Rodrigues and Mariana Henriques

CEB, Centre of Biological Engineering, LIBRO – Laboratório de Investigação em Biofilmes Rosário Oliveira, University of Minho, Braga, Portugal

ABSTRACT ARTICLE HISTORY Candidiases are the most recurrent fungal infections, especially among immunosuppressed patients. Received 9 January 2018 Although Candida albicans is still the most widespread isolated species, non-Candida albicans Accepted 28 April 2018 Candida species have been increasing. The goal of this work was to determine the susceptibility of KEYWORDS C. glabrata biofilms to echinocandins and to evaluate their effect on the biofilm matrix composition, Candida glabrata; comparing the results with other Candida species. Drug susceptibilities were assessed through echinocandin; biofilm; the determination of minimum inhibitory concentration (MIC), minimum fungicidal concentration matrix; antifungal drug; (MFC) and minimum biofilm eradication concentration (MBEC) of caspofungin (Csf) and micafugin resistance (Mcf). The β-1,3 glucans content of the matrices was assessed after contact with the drugs. The data suggest that, generally, after contact with echinocandins, the concentration of β-1,3 glucans increased. These adjustments in the matrix composition of C. glabrata biofilms and the chemical differences between Csf and Mcf, seem responsible and may determine the effectivity of the drug responses.

Introduction of exopolymeric compounds secreted by sessile cells, with all providing protection against environmental challenges Infections caused by members of the genus Candida (Pierce et al. 2017). Infections caused by biofilms are com- (candidiasis) have been increasing in recent decades and plicated due to inducible gene networks encoding different becoming more difficult to eradicate. There are about 15 proteins that confer tolerance or resistance to many of the differentCandida species that cause infections in humans, available antifungal drugs (Ramage et al. 2012). and Candida glabrata is one of the most common (Pappas Chemically, echinocandins are cyclic lipo-hexapeptides 2006; Pappas et al. 2015). Candidiasis generally occurs due with modified N-linked acyl lipid side chains (Chang to the unbalanced use of immunosuppressive drugs, broad et al. 2017), biosynthesised by diverse members of the spectrum antibiotics and the widespread use of indwell- Ascomycota (fungi) on non-ribosomal peptide synthase ing medical devices, but also to the growth of immuno- complexes (Emri et al. 2013). The first echinocandin with genic diseases, the upsurge of endocrine disorders, and antimycotic activity was discovered in the 1970s, and the ageing and the increase in the patient population since, over 20 natural echinocandins have been isolated (Douglas 2003; Li et al. 2007; Garcia-Cuesta et al. 2014; (Emri et al. 2013). By disturbing fungal cell wall synthe- Silva et al. 2017). Each Candida species has distinctive sis through a non-competitive inhibition of β-1,3-glucan virulence factors, antifungal susceptibilities, and defined synthesis, these drugs weaken the cell wall, break down epidemiologies (Pappas et al. 2015). The aptitude of these the cellular integrity and, finally, induce cell lysis (Debono organisms to form biofilms is a specific virulence feature and Gordee 1994; Perlin et al. 2007). Due to this mech- which allows tissue attachment following infection of the anism of action the echinocandins (which include anid- host (McCall and Edgerton 2017). Biofilms are commu- ulafungin, caspofungin and micafungin) are generally nities of microorganisms embedded in an extracellular well tolerated, avoiding the overlapping toxicities and matrix (Costerton et al. 1995; Donlan and Costerton drug–drug interactions with mammalian cells which are 2002), which confers significant resistance to antifungal observed with the azoles and the polyenes (Wiederhold therapy and intense host immune responses (Fonseca et and Lewis 2003; Wiederhold et al. 2007; Chang et al. al. 2014; Rodrigues et al. 2014). This matrix is composed 2017). The high clinical efficacy in the non-neutropenic

CONTACT Mariana Henriques [email protected] © 2018 Informa UK Limited, trading as Taylor & Francis Group 570 C. F. RODRIGUES ET AL. patient population, in patients with moderate to severe cavity (C. glabrata AE2 and D1), urinary tract (C. glabrata illness, and in patients with pre-azole exposure (Pappas 562123 and 513100), vaginal tract (C. glabrata 534784 and et al. 2015; Perlin 2015b), resulted in echinocandins being 585626); four reference strains were from the American recommended as the first-line antifungal agents to treat Type Culture Collection (C. glabrata ATCC2001, C. invasive candidiasis, especially Candida glabrata, due albicans SC5314 (ATCCMYA2876), C. parapsilosis to its innate high azole resistance (Perlin 2007; Pappas ATCC22019 and C. tropicalis ATCC750). The identity of et al. 2015). Although these antifungal drugs are active all isolates was confirmed using CHROMagarCandida against most important Candida species, in which they (CHROMagar, Paris, France) and by PCR-based sequenc- display in vitro fungicidal activity (Barchiesi et al. 2005), ing using specific primers ITS1( and ITS4) against the in critically ill patients it is recognised that the achieve- 5.8s subunit gene reference. Genomic DNA was extracted ment of their pharmacodynamic and pharmacokinetic following previously described procedures (Williams et al. targets show a large inter-individual variability (Chang 1995). The PCR products were sequenced using the ABI- et al. 2017). In Europe, micafungin (Mcf) is approved for PRISM Big Dye terminator cycle sequencing kit (Perkin use in paediatric patients of any age including neonates, Elmer, Applied Biosystems, Warrington, UK). while caspofungin (Csf) is approved for use in paediatric patients ≥ 1 year old, since there are insufficient data Growth conditions regarding its use in those <1 year old (Tragiannidis et al. 2012; Viscoli et al. 2014). The use of Csf and Mcf is lim- For each experiment, strains were subcultured on ited by the necessity for a once-daily intravenous dosage Sabouraud dextrose agar (SDA) (Merck, Darmstadt, regimen, lack of an oral formulation and a limited spec- Germany) for 24 h at 37°C. Cells were then inoculated in trum (Barchiesi et al. 2005; Pappas et al. 2015; McCarty Sabouraud dextrose broth (SDB) (Merck) and incubated and Pappas 2016), but both echinocandins still show very for 18 h at 37°C, under agitation at 120 rpm. After incu- good in vitro activity against clinically relevant isolates of bation, the cells were harvested by centrifugation, 3,000 g Candida species (Pappas et al. 2007; Spreghini et al. 2012; for 10 min, at 4°C and washed twice with phosphate Kohno et al. 2013; Perlin 2015a; Chapman et al. 2017). In buffer saline (PBS, 0.1 M, pH=7.5; NaCl 0.8%, KCl 0.02%, trials involving adult and paediatric patients with invasive K2HPO4 0.02%, NaHPO4.12H2O 0.285%). Pellets were and oesophageal candidiasis, Mcf has been shown to be then suspended in RPMI-1640 (pH=7, Sigma-Aldrich, non-inferior to intravenous Csf, intravenous fluconazole St Louis, MO, USA) and the cellular density was adjusted or liposomal amphotericin B. The tolerability profile of to 1×105 cells ml−1, using a Neubauer counting chamber. Csf and Mcf are, in general, similar to fluconazole and are better tolerated than liposomal amphotericin B or oral Antifungal drugs itraconazole (Scott 2012). The biofilm matrices ofCandida species have a strong Csf and Mcf were kindly provided by MSD® and Astellas®, net of exopolymers, providing protection against physi- respectively. Aliquots of 5,000 mg l−1 were prepared using cal and chemical environmental attack, such as by drugs. dimethyl-sulfoxide (DMSO). The final concentrations These polymers make it difficult for drugs to diffuse into used were prepared with RPMI-1640 (Sigma-Aldrich) the biofilm cells, which make the biofilms recalcitrant to for both drugs. antifungals (Al-Fattani and Douglas 2004; Rodrigues, Gonçalves et al. 2017; Dominguez et al. 2018). Hence, Antifungal susceptibility tests the goal of this work was to evaluate seven C. glabrata isolates, by comparison with C. albicans, C. parapsilosis All the antifungal susceptibility tests were performed and C. tropicalis, regarding their susceptibility in the using the microdilution method, in accordance to the planktonic and biofilm form to Csf and Mcf, and the bio- European Committee on Antimicrobial Susceptibility chemical variations induced in the composition of the Testing (EUCAST) guidelines (EUCAST [date unknown]; matrices after the drug exposure, in order to relate these Arendrup et al. 2008). matrix variations with drug effectiveness. Minimum inhibitory concentrations (MICs) Material and methods The inoculum was prepared by suspending five distinct colonies, ≥1 mm diameter from 24 h cultures, in at least Organisms 3 ml of sterile distilled water. Then, the inoculum was A total of 10 strains were used in the course of this study. suspended by vigorous shaking on a vortex mixer for 15 Six clinical isolates of C. glabrata from Hospital Escala s and the cell density was adjusted to the density of a 0.5 Braga in Portugal were recovered from different sites: oral McFarland standard and adding sterile distilled water as BIOFOULING 571 required, giving a yeast suspension of 1–5 × 106 CFU ml−1 2017). As described previously, after biofilm formation for (colony forming units). A working suspension was pre- a total of 48 h, the medium was aspirated and non-adher- pared by a dilution of the standardised suspension in ster- ent cells removed by washing the biofilms with PBS. Then, ile distilled water to yield 1–5 × 105 CFU ml−1. The 96-well biofilms were fixed with 200 μl of methanol, which was plate was prepared with 100 μl of cell suspension and 100 μl removed after 15 min of contact. The microtitre plates of each antifungal agent (Csf: 0.001 to 0.075 mg l−1; Mcf: were allowed to dry at room temperature, and 200 μl of 0.002 to 0.2 mg l−1 – 2x concentrated) and incubated at CV (1% v v–1) were added to each well and incubated 37°C, during 18–48 h. Controls without antifungal agents for 5 min. The wells were then gently washed twice with were also performed (positive control: working solution sterile, ultra-pure water and 200 μl of acetic acid (33% of cells and RPMI-1640; negative control: sterile distilled v v–1) were added to release and dissolve the stain. The water and RPMI-1640). Finally, the results were deter- absorbance of the obtained solution was read in triplicate mined spectrophotometrically at 530 nm and the MIC was in a microtitre plate reader (Bio-Tek Synergy HT, Izasa, considered to be the value that inhibited 50%, compared Lisbon, Portugal) at 570 nm. The results were presented to the controls (according to EUCAST guidelines). as absorbance per unit area (Abs cm−2).

Minimum fungicidal concentration (MFC) Biofilm structure visualisation In addition to the previous step, 20 μl of each cell suspen- In order to examine the structure of biofilms, after biofilm sion treated with Csf and Mcf were recovered to a new formation in the presence or absence of both drugs, they well and serial decimal dilutions in PBS were plated onto were observed by scanning electron microscopy (SEM). SDA. Agar plates were incubated for 24 h at 37°C, and the For that, biofilms formed as described above were dehy- total number of CFUs was determined. The results were drated with ethanol (using 70% ethanol for 10 min, 95% −2 presented as Log10 CFU per unit area (Log10 CFU cm ) ethanol for 10 min and 100% ethanol for 20 min) and (Rodrigues and Henriques 2017). air dried for 20 min. Samples were kept in a desiccator until further analysis. Prior to observation, the base of the Minimum biofilm eradication concentration (MBEC) wells was mounted onto aluminium stubs, sputter coated Standardised cell suspensions (200 μl) were placed into with gold and observed with an S-360 scanning electron selected wells of 96-well polystyrene microtitre plates microscope (Leo, Cambridge, MA, USA) (Rodrigues and (Orange Scientific, Braine-l’Alleud, Belgium). RPMI-1640 Henriques 2017). was used without cells, but without antifungal agent, as a negative control. As positive control, cell suspensions were Biofilm matrix composition evaluation tested with RPMI-1640, without the antifungal agent. At Extraction method: biofilms were formed in 24-well poly- 24 h, 100 μl of RPMI-1640 were removed and an equal styrene microtitre plates (Orange Scientific) (Rodrigues, volume of fresh RPMI-1640 plus the respective antifungal Gonçalves et al. 2017). For this, 1,000 μl of yeast cell sus- concentration was added (Csf: 0.5 to 3 mg l−1; Mcf: 3 to pension (1 × 105 cells ml−1 in RPMI-1640 or RPMI-1640 17 mg l−1; 2× concentrated). The plates were incubated with Csf and Mcf at concentrations corresponding to the at 37°C for more 24 h, a total of 48 h, at 120 rpm. The MBEC of each species/strain) were added to each well number of cultivable cells on biofilms was determined by and the biofilms were treated as described previously the enumeration of CFUs. For that, after the period of bio- (Rodrigues, Gonçalves et al. 2017). After 24 h, 500 μl of film formation, all medium was aspired and the biofilms RPMI-1640 medium were removed and an equal volume washed once with 200 μl of PBS to remove non-adherent of fresh RPMI-1640 with or without the antifungal agents cells. Then, biofilms were scraped from the wells and the was added. After another 24 h, the biofilms were scraped suspensions were vigorously vortexed for 2 min to disag- from the 24-well plates, resuspended in ultra-pure water, gregate cells from the matrix. Serial decimal dilutions in sonicated (Ultrasonic Processor, Cole-Parmer, Vernon PBS were plated on SDA and incubated for 24 h at 37°C. Hills, IL, USA) for 30 s at 30 W, and then the suspension The results were presented as total CFUs per unit area vortexed for 2 min. The suspension was centrifuged at −2 (Log10 CFU cm ) (Rodrigues, Gonçalves et al. 2017). 5,000 g for 5 min at 4°C. The supernatant (matrix) was filtered through a 0.2 μm nitrocellulose filter and dried at 37°C until a constant dry biofilm weight was reached. The Biofilm analysis pellets (sessile yeast cells) were discarded. Biofilm total biomass quantification – crystal violet β-1,3 glucan concentration determination: β-1,3 glucan staining concentrations were determined using Glucatell® kit (Cape Total biofilm biomass was quantified by crystal violet Cod®, East Falmouth, MA, USA). The values were nor- (CV) staining in 96-well plates (Rodrigues and Henriques malised per pg μg−1 of β-1,3 glucans/total polysaccharide 572 C. F. RODRIGUES ET AL. content (evaluated by the phenol-sulfuric procedure using Table 1. MIC, MFC and MBEC values (mg l−1) for Csf and Mcf of C. glucose as standard (DuBois et al. 1956). glabrata, C. albicans, C. parapsilosis and C. tropicalis strains. Determi- Caspo- Origin Species/strain nation fungin Micafungin Statistical analysis Reference C. glabrata ATCC2001 MIC 0.064 0.032 MFC 0.10 0.20 All experiments were repeated three times in at least three MBEC 2.5–3 16–17 independent assays. The results were compared using one- Oral cavity AE2 MIC ≤0.003 0.002 MFC 0.003 0.06 way and two-way ANOVA, Tukey’s and Dunnett’s multiple MBEC 0.5–1 7–8 comparisons test, using GraphPad® Prism® 7 software (San D1 MIC 0.003 0.002 MFC 0.003 0.06 Diego, CA, USA). All tests were performed with a confi- MBEC 2.5–3 3–3.5 dence level of 95%. Urinary 562123 MIC 0.001 0.01 tract MFC 0.001 0.10 MBEC 0.5–1 16–17 513100 MIC 0.008 0.01 Results MFC 0.008 0.06 MBEC 2–2.5 16 Planktonic and biofilm susceptibility to Csf and Mcf Vaginal 534784 MIC ≤0.001 0.002 tract MFC 0.05 0.06 Table 1 shows the MIC, MFC and MBEC determined for MBEC 2.5–3 5.5–6 585626 MIC 0.003 0.01 all the strains used in this study. The values were species/ MFC 0.003 0.06 strain dependent, but, in general, MFC and MBEC of MBEC 2.5 5–5.5 Mcf were higher than those from Csf. This was especially Reference C. albicans SC5314 MIC 0.016 ≤0.016 MFC 0.016 0.06 noticeable for the concentration needed to eradicate the MBEC 2.5–3 3.5 biofilm (MBEC), which was sometimes 5–6 times higher Reference C. parapsilosis MIC 0.016 >0.032 ATCC22019 MFC 0.08 0.2 than the MFC. Mostly, all C. glabrata strains were demon- MBEC 2.5 16 strated to have biofilm resistance profiles similar to the Reference C. tropicalis ATCC750 MIC 0.008 0.002 MFC 0.016 0.06 other Candida species. MBEC 2.5 16 Note: Bold indicates MBEC values. Biofilm reduction capacity of Csf and Mcf

CV staining was used to evaluate the biomass reduction in Table 2. Percentage biomass reduction of C. glabrata, C. albicans, the Candida biofilms, after 24 h contact with Csf or Mcf C. parapsilosis and C. tropicalis strains after caspofungin and mi- (Table 2). Although both agents showed a good capac- cafungin contact. ity in reducing the biomass, Csf had a higher capacity % Biomass reduction (p-value) than Mcf. Four of the seven C. glabrata strains (585626, Species/strain Caspofungin Micafungin 534784, D1 and ATCC2001) had the lowest percentage C. glabrata ATCC2001 82.96 (**) 19.1 (*) AE2 79.71 (***) 39.28 (**) biomass reduction for Csf and Mcf, indicating a more tol- D1 69.71 (*) 23.84 (ns) erant profile of this species, comparing to the other species 562123 85.16 (**) 70.47 (**) of Candida. On the other hand, C. albicans SC5314 (p < 513100 88.47 (****) 66.59 (****) 534784 73.38 (*) 24.46 (ns) 0.001 for Csf and Mcf) was demonstrated to have the most 585626 77.85 (**) 53.63 (**) marked biomass reductions. C. albicans SC5314 92.33 (***) 82.13 (***) C. parapsilosis ATCC22019 78.55 (***) 38.86 (**) C. tropicalis ATCC750 84.20 (**) 53.49 (*) Biofilm structure Notes: ns, non-significant. The concentrations applied in each species/strain were those determined by SEM images (Figure 1A and B) confirmed that all Candida the MBECs. Overall ANOVA p<0.05 and post hoc Dunnett’s comparison test: *p < 0.05; **p < 0.001; ***p < 0.0005; ****p < 0.0001 strains had a good capacity for biofilm production, especially C. glabrata AE2, C. glabrata D1 (biofilms formed in a long continuous carpet (Hawser and Douglas 1995; with large amounts of extracellular material (Heffner and Fonseca et al. 2014; Rodrigues and Henriques 2017), C. Franklin 1978; Rodrigues and Henriques 2017; Silva et al. albicans SC5314 (biofilm presenting high hyphal quantity 2017). After Csf and Mcf contact, the SEM images con- and entanglement (Uppuluri et al. 2010; Rodrigues and firmed the CV results for all species/strains (Figure1 B). Henriques 2017) and C. parapsilosis ATCC22019 (con- A biofilm reduction was observed in the presence of both tinuous biofilm carpet with clumped blastospores (Taff drugs, but especially with Csf, and the biofilm cells pre- et al. 2012; Rodrigues and Henriques 2017). C. tropica- sented a concave aspect and appear to have a reduction in lis ATCC750 biofilm can be described as chains of cells the extracellular matrix. However, the extracellular matrix BIOFOULING 573

(A)

Figure 2. β-1,3 glucans concentration/polysaccharides content (pg μg−1) in 48-h biofilm matrices ofC. glabrata, C. albicans (Ca), C. parapsilosis (Cp) and C. tropicalis (Ct). Overall ANOVA p<0.05 ANOVA and post hoc Tukey’s comparisons test: *p<0.05; **p<0.001; ***p<0.0005; ****p<0.0001).

of the different biofilms in the presence of the drugs seems to be different, which may be explained by variations in β-1,3 glucans, as was evaluated subsequently.

Matrix composition after Csf and Mcf contact The results from the determination of the β-1,3 glucans concentration showed that, compared to the control groups and with some exceptions, the β-1,3 glucans generally tend to statistically decrease in the biofilm matrices of all Candida species, after contact with Csf and especially with Mcf (Figure 2). This reduction was not so promi- nent in C. glabrata, when compared to the other species (B) (Figure 2).

Discussion Systemic candidiasis is a growing problem in the hospitals worldwide (Cataldi et al. 2016; Rodrigues, Rodrigues et al. 2017), with high morbidity and mortality rates and ele- vated economic costs (Pfaller and Castanheira 2016; Silva et al. 2017). The present study evaluated the efficacy of two of the most used echinocandins – the first-line antifungal agents to treat systemic candidiasis (Pappas et al. 2015; McCarty and Pappas 2016) – of seven C. glabrata strains compared with the reference strains C. albicans, C. parapsilosis and C. tropicalis. Figure 1. SEM observations of biofilmsC. glabrata (A), C. albicans, Considering MIC values of Mcf, EUCAST guide- C. parapsilosis and C. tropicalis (B) strains/species grown without lines indicate breakpoints of 0.032 mg l−1 for C. glabrata, drugs (control) and after caspofungin and micafungin contact. 0.016 mg l1 for C. albicans, 0.002–2 mg l−1 for C. parapsilo- The concentrations applied to each species/strain were those sis. For C. tropicalis, the values are defined as 1–2 twofold determined by the MBECs. The arrows show disrupted biofilms and damaged Candida cells. Magnification: 1,000×. Scale dilution steps higher than for C. albicans and C. glabrata, bar=20 μm. but EUCAST attest that there is insufficient evidence to 574 C. F. RODRIGUES ET AL. indicate whether the wild-type population can be consid- have been indicated previously in studies with catheters ered susceptible to Mcf (EUCAST). For MIC Csf values, (Falcone et al. 2009; Dutronc et al. 2010; Williams and EUCAST breakpoints have not yet been established, due to Ramage 2015). Mcf did not reveal a complete eradication significant inter-laboratory variation in the ranges for this of the biofilm, as it has been reported for Csf (Cateau et al. drug (EUCAST). Possibly because of this, the authors had 2008, 2011). This conclusion was particularly noticed difficulty in defining these parameters. The occurrence in C. glabrata (four of the seven strains had the lowest of paradoxical growth of the isolates of Candida species, Mcf biomass percentage reduction), which can be related also enhanced the difficulties in MIC determination. The to the usual antifungal resistance profile of this species paradoxical effect is recognised as a robust mechanism (Jabra-rizk et al. 2004; Farmakiotis et al. 2014; Perlin et of antifungal resistance. It is a resurgence of growth at al. 2015; Silva et al. 2017). Interestingly, the efficacy of the drug concentrations above the MIC and connected to an biomass reduction by Csf was not always related to the use increase in chitin biosynthesis whereby susceptible cells of the highest drug concentration. In some cases, a high show growth at very high drug levels. Candida isolates Csf concentration would lead to a low drug activity and, display paradoxical effect more frequently when grown thus, to a lower biomass reduction, revealing a probable as biofilms compared to the planktonic form (Gow et al. paradoxical effect (Wiederhold 2009; Perlin 2014; Gil- 2007; Gil-Alonso et al. 2015, 2016; Scorzoni et al. 2017). Alonso et al. 2015; Scorzoni et al. 2017). By applying an The results, however, confirmed that all species/strains adequate concentration for the strain (sometimes a lower were Mcf susceptible (Table 1). EUCAST guidelines state drug dose), this effect was not detected, and the biomass that isolates that are susceptible to Mcf should be con- was effectively reduced. sidered susceptible to Csf (EUCAST), hence all species/ In order to evaluate the extent of the antifungal effect strains were considered susceptible to this drug (Table of Csf and Mcf on the biofilms, SEM images were taken 1). Although Mcf concentrations were, in most cases, (Figure 1A and B). The controls revealed that all Candida higher than Csf concentrations, in general, the MIC, MFC species had a good capacity for biofilm production. After and MBEC values were species/strain dependent. C. gla- Csf and Mcf contact (Figure 1B), and particularly with Csf, brata strains were revealed to have similar MIC, MFC a strong biofilm reduction, was observed (arrows in the and MBEC profiles toC. albicans SC5314, C. tropicalis Figure 1). The biofilm cells seemed to be disrupted and ATCC750 and C. parapsilosis ATCC22019. Not surpris- presented a concave aspect (arrows in Figure 1), endors- ingly, the MBEC results evidenced the higher resistant ing the CV results (Table 2). This disruption and change profile of biofilm cells to antifungal agents than plank- in the cell configuration is expected to result from the tonic cells (Lewis et al. 2002; Al-Fattani and Douglas 2004; mechanism of action of the echinocandins (non-com- Ferrari et al. 2011; Grandesso et al. 2012; De Luca et al. petitive inhibition of β-1,3-glucan synthesis) (Rodrigues, 2012). In general, C. glabrata MIC values were shown to be Rodrigues et al. 2017; Scorzoni et al. 2017), which affects closer, but slightly lower than those reported in the litera- the cell wall and the matrix composition. Fungal cell wall ture for some strains (Scott 2012; Naicker et al. 2016) and polysaccharides are significant constituents of the biofilm that all the strains were more susceptible to both echino- matrix of Candida species (Chandra et al. 2001; Kuhn candins than C. glabrata ATCC2001 (EUCAST). Yet, and et al. 2002). The biofilm matrix composition ofC. gla- as supported by the reports, C. parapsilosis showed higher brata, C. albicans, C. parapsilosis and C. tropicalis strains MIC values than the other Candida species (Arendrup et with and without Csf and Mcf contact were evaluated in al. 2012; Arendrup and Pfaller 2012; Arendrup and Perlin terms of β-1,3 glucan content (Figure 2). To the authors’ 2014; Pham et al. 2014; Gil-Alonso et al. 2015). It is known knowledge, this is the first report describing such content that some Candida species have naturally occurring pol- of the biofilm matrices ofCandida species in contact with ymorphisms in their FKS genes, which strongly reduce these two echinocandins. The β-1,3 glucans are a group their susceptibility to echinocandin drugs (Arendrup et al. of specific polysaccharides from the cell walls ofCandida 2012; Perlin 2015a). Moreover, the C. parapsilosis family species that are also recognised as major constituents of (C. orthopsilosis and C. metapsilosis) and Candida guillier- the biofilm matrices of this genus (Chandra et al.2001 ; mondii, normally also have higher MIC values compared Kuhn et al. 2002). Regarding the controls (non-treated to other susceptible Candida species (Pfaller et al. 2008, biofilms), C. albicans SC5314 was the species that had a 2010; Tortorano et al. 2013; Perlin 2015b). lower quantify of β-1,3 glucans in the matrix and C. par- The results of CV staining indicated, altogether, an apsilosis ATCC22019, C. glabrata 534784 and C. glabrata effective reduction in the biomass of the species/strains 562123 were revealed to have the highest content of this after echinocandin exposure (Table 2). Yet, Csf showed glucan, which could be one of the probable mechanisms a higher biomass reduction capacity than Mcf, with a for the resistant profile of the biofilms (Figure 2). After minimal reduction of 70% of the biofilms. Similar results adding Csf and Mcf to the pre-formed biofilms, the results BIOFOULING 575 from the determination of the β-1,3 glucan concentra- Funding tions (Figure 2) demonstrated that these compounds are, This study was supported by the Portuguese Foundation for in general, statistically significantly likely to reduce in the Science and Technology (FCT) under the scope of the strategic biofilm matrices of C. glabrata, and in the other Candida funding of UID/BIO/04469/2013 unit and COMPETE 2020 species. Exceptions were noted only after the addition [POCI-01–0145-FEDER-006684] and BioTecNorte operation of Csf to C. glabrata ATCC2001, C. glabrata D1 and C. [NORTE-01–0145-FEDER-000004] funded by the European tropicalis ATCC750. It is also important to note that the Regional Development Fund under the scope of Norte2020 - Programa Operacional Regional do Norte, Célia F. Rodrigues’ initial concentration of β-1,3 glucans was very dependent [SFRH/BD/93078/2013] PhD grant and M. Elisa Rodrigues on the strain and that, excepting C. glabrata 534784, the [SFRH/BPD/95401/2013] post-doctoral grant. variations between the strains and the Mcf groups were less pronounced than those detected between the strains ORCID and Csf groups. Another source of variability in the drug response Célia F. Rodrigues http://orcid.org/0000-0001-8633-2230 might be the chemical differences between the two echino- Maria Elisa Rodrigues http://orcid.org/0000-0001-8823- candin molecules. Both Csf and Mcf have a cyclic peptide 9494 structure with a N-aryl group but with different patterns of Mariana Henriques http://orcid.org/0000-0003-0317-4877 hydroxylations and amino groups (R2 to R4). The N-aryl side chain (position R1) plays a critical role in the potency References and toxicity and is the main point for chemical modifica- tion of the echinocandin analogues (Debono and Gordee Al-Fattani M, Douglas LJ. 2004. Penetration of Candida biofilms by antifungal agents. Antimicrob Agents Chemother. 1994; Wiederhold and Lewis 2003). Csf and Mcf have also 48:3291–3297. doi:10.1128/AAC.48.9.3291-3297.2004. variations in R2, R3 and R4 positions: Csf is more hydrox- Arendrup MC, Arikan S, Barchiesi F, Bille J, Dannaoui E, ylated and has more amino groups, while Mcf has more Denning DW, Donnelly JP, Fegeler W, Moore C, Richardson aryl groups. It is documented that the replacement of the M, et al. 2008. EUCAST Technical Note on the method for linoleoyl side chain with aryl side chains of low lipophilic- the determination of broth dilution minimum inhibitory ity, nonlinear configuration or chains switched with highly concentrations of antifungal agents for conidia – forming moulds. ESCMID Tech Notes. 14:982–984. polar groups end in loss of antifungal activity, which sug- Arendrup M, Perlin D, Jensen R, Howard S, Goodwin J, Hopec gests that planar, non-polar substitutions are critical for the W. 2012. Differentialin vivo activities of anidulafungin, antifungal activity (Debono and Gordee 1994; Wiederhold caspofungin, and micafungin against Candida glabrata and Lewis 2003). These modifications were performed on isolates with and without FSK resistance mutations. Antim the Mcf molecule and are also a probable explanation for Agents Chemoter. 56:2435–2442. Arendrup MC, Perlin DS. 2014. Echinocandin resistance: an the differences found, in this work, for the drug activity in emerging clinical problem? Curr Opin Infect Dis. 27:484–492. the species and strains studied (higher MBEC values and Arendrup MC, Pfaller Ma. 2012. Caspofungin Etest matrix variations, when compared to Csf). susceptibility testing of Candida species: risk of This report shows a general parallelism in the efficacy of misclassification of susceptible isolates ofC. glabrata and the Csf and Mcf susceptibilities of C. glabrata in comparison C. krusei when adopting the revised CLSI caspofungin to C. albicans, C. parapsilosis and C. tropicalis with respect breakpoints. Antimicrob Agents Chemother. 56:3965–3968. doi:10.1128/AAC.00355-12. to planktonic cells (MIC/MFC). It seems plausible to say Barchiesi F, Spreghini E, Tomassetti S, Arzeni D, Giannini D, that the external alterations in the matrix composition of Scalise G. 2005. Comparison of the fungicidal activities of the Candida strains and the chemical alterations in the Csf caspofungin and amphotericin B against Candida glabrata. and Mcf molecules can partially explain and determine the Antimicrob Agents Chemother. 49:4989–4992. doi:10.1128/ effectiveness response to these two drugs of biofilm infec- AAC.49.12.4989-4992.2005. Cataldi V, Di Campli E, Fazii P, Traini T, Cellini L, Di Giulio tions of Candida species and, particularly, C. glabrata. M. 2016. Candida species isolated from different body sites and their antifungal susceptibility pattern: cross-analysis Acknowledgements of Candida albicans and Candida glabrata biofilms. Med Mycol:myw126. doi:10.1093/mmy/myw126. The authors would like to acknowledge MSD® and Astellas® for Cateau E, Berjeaud J-M, Imbert C. 2011. Possible role of azole the kind donation of Caspofungin and Micafungin, respectively. and echinocandin lock solutions in the control of Candida biofilms associated with silicone. Int J Antimicrob Agents. 37:380–384. doi:10.1016/j.ijantimicag.2010.12.016. Disclosure statement Cateau E, Rodier M-H, Imbert C. 2008. In vitro efficacies of caspofungin or micafungin catheter lock solutions on All the authors declare there is no financial/personal interest or Candida albicans biofilm growth. J Antimicrob Chemother. belief that could affect their objectivity. 62:153–155. doi:10.1093/jac/dkn160. 576 C. F. RODRIGUES ET AL.

Chandra J, Kuhn D, Mukherjee P, Hoyer L, McCormick T, Fonseca E, Silva S, Rodrigues C, Alves C, Azeredo J, Henriques Ghannoum M. 2001. Biofilm formation by the fungal M. 2014. Effects of fluconazole onCandida glabrata biofilms pathogen Candida albicans: development, architecture, and and its relationship with ABC transporter gene expression. drug resistance. J Bacteriol. 183:5385–5394. doi:10.1128/ Biofouling. 30:447–457. doi:10.1080/08927014.2014.886108. JB.183.18.5385-5394.2001. Garcia-Cuesta C, Sarrion-Pérez M-G, Bagán JV, Sarrion-Perez Chang CC, Slavin MA, Chen SCAC-A. 2017. New MG, Bagan JV. 2014. Current treatment of oral candidiasis: a developments and directions in the clinical application of the literature review. J Clin Exp Dent. 6:e576–e582. doi:10.4317/ echinocandins. Arch Toxicol. 91:1613–1621. doi:10.1007/ jced.51798. s00204-016-1916-3. Gil-Alonso S, Jauregizar N, Cantón E, Eraso E, Quindós Chapman B, Slavin M, Marriott D, Halliday C, Kidd SE, Arthur G. 2015. In Vitro fungicidal activities of anidulafungin, I, Bak N, Heath CH, Kennedy K, Morrissey CO, et al. 2017. caspofungin, and micafungin against Candida glabrata, Changing epidemiology of candidaemia in Australia. J Candida bracarensis, and Candida nivariensis evaluated by Antimicrob Chemother. 72:1103–1108. time-kill studies. Antimicrob Agents Chemother. 59:3615– Costerton JW, Lewandowski Z, Caldwell DE, Korber DR, 3618. doi:10.1128/AAC.04474-14. Lappin-Scott HM. 1995. Microbial biofilms. Annu Gil-Alonso S, Jauregizar N, Ortega I, Eraso E, Suárez E, Rev Microbiol. 49:711–745. doi:10.1146/annurev. Quindós G. 2016. In vitro pharmacodynamic modelling mi.49.100195.003431. of anidulafungin against Candida spp. Int J Antimicrob De Luca C, Guglielminetti M, Ferrario A, Calabrὸ M, Casari E. Agents. 47:178–183. doi:10.1016/j.ijantimicag.2015.12.011. 2012. Candidemia: species involved, virulence factors and Gow NAR, Netea MG, Munro CA, Ferwerda G, Bates S, Mora- antimycotic susceptibility. New Microbiol. 35:459–468. Montes HM, Walker L, Jansen T, Jacobs L, Tsoni V, et al. 2007. Debono M, Gordee RSS. 1994. Antibiotics that inhibit fungal Immune recognition of Candida albicans beta-glucan by cell wall development. Annu Rev Microbiol. 48:471–497. dectin-1. J Infect Dis. 196:1565–1571. doi:10.1086/525005. doi:10.1146/annurev.mi.48.100194.002351. Grandesso S, Sapino B, Mazzuccato S, Solinas M, Bedin M, Dominguez E, Zarnowski R, Sanchez H, Covelli AS, Westler D’Angelo M, Gion M. 2012. Study on in vitro susceptibility WM, Azadi P, Nett J, Mitchell AP, Andes DR. 2018. of Candida spp. isolated from blood culture. Infect Med. Conservation and divergence in the Candida species biofilm 20:25–30. matrix mannan-glucan complex structure, function, and Hawser SP, Douglas LJ. 1995. Resistance of Candida albicans genetic control. MBio. 9:e00451–18. biofilms to antifungal agentsin vitro. Antimicrob Agents Donlan R, Costerton J. 2002. Biofilms: survival mechanisms Chemother. 39:2128–2131. doi:10.1128/AAC.39.9.2128. of clinically relevant microorganism. Clin Microbiol Rev. Heffner D, Franklin W.1978 . Endocarditis caused by Torulopsis 15:167–193. doi:10.1128/CMR.15.2.167-193.2002. glabrata. Am J Clin Pathol. 70:420–423. doi:10.1093/ Douglas LJ. 2003. Candida biofilms and their role in ajcp/70.3.420. infection. Trends Microbiol. 11:30–36. doi:10.1016/S0966- Jabra-rizk MA, Falker WA, Meiller TF. 2004. Fungal biofilms 842X(02)00002-1. and drug resistance. Emerg Infect. 10:19. DuBois M, Gilles K, Hamilton J, Rebers P, Smith F. 1956. Kohno S, Izumikawa K, Yoshida M, Takesue Y, Oka S, Kamei Colorimetric method for determination of sugars and related K, Miyazaki Y, Yoshinari T, Kartsonis NA, Niki Y. 2013. A substances. Anal Chem. 28:350–356. doi:10.1021/ac60111a017. double-blind comparative study of the safety and efficacy Dutronc H, Dauchy FA, Cazanave C, Rougie C, Lafarie-Castet of caspofungin versus micafungin in the treatment of S, Couprie B, Fabre T, Dupon M. 2010. Candida prosthetic candidiasis and aspergillosis. Eur J Clin Microbiol Infect infections: case series and literature review. Scand J Infect Dis. 32:387–397. doi:10.1007/s10096-012-1754-z. Dis. 42:890–895. doi:10.3109/00365548.2010.498023. Kuhn D, Chandra J, Mukherjee P, Ghannoum M. 2002. Emri T, Majoros L, Tóth V, Pócsi I. 2013. Echinocandins: Comparison of biofilms formed by Candida albicans and production and applications. Appl Microbiol Biotechnol. Candida parapsilosis on bioprosthetic surfaces. Infect 97:3267–3284. doi:10.1007/s00253-013-4761-9. Immun. 70:878–888. doi:10.1128/IAI.70.2.878-888.2002. EUCAST. [date unknown]. Breakpoint tables for Lewis R, Kontoyiannis D, Darouiche R, Raad I, Prince R. 2002. interpretation of MICs, Version 8.1, valid from 2017-03- Antifungal activity of amphotericin B, fluconazole, and 01. http://www.eucast.org/fileadmin/src/media/PDFs/ voriconazole in an in vitro model of Candida catheter-related EUCAST_files/AFST/Clinical_breakpoints/Antifungal_ bloodstream infection. Antimicrob Agents Chemother. breakpoints_v_8.1_March_2017.pdf 46:3499–3505. doi:10.1128/AAC.46.11.3499-3505.2002. Falcone M, Barzaghi N, Carosi G, Grossi P, Minoli L, Ravasio Li L, Redding S, Dongari B. 2007. Candida glabrata , an V, Rizzi M, Suter F, Utili R, Viscoli C, et al. 2009. Candida Emerging Oral Opportunistic Pathogen. J Dent Res. 86:204– infective endocarditis. Medicine (Baltimore). 88:160–168. 215. doi:10.1177/154405910708600304. doi:10.1097/MD.0b013e3181a693f8. McCall A, Edgerton M. 2017. Real-time approach to flow cell Farmakiotis D, Tarrand JJJ, Kontoyiannis DPDP. 2014. Drug- imaging of Candida albicans biofilm development. J Fungi. resistance Candida glabrata infection in cancer patients. 3:13. doi:10.3390/jof3010013. Emerg Infect Dis. 20:11–26. McCarty TP, Pappas PG. 2016. Invasive candidiasis. Infect Dis Ferrari S, Sanguinetti M, De Bernardis F, Torelli R, Posteraro Clin North Am. 30:103–124. B, Vandeputte P, Sanglard D. 2011. Loss of mitochondrial Naicker SD, Magobo RE, Zulu TG, Maphanga TG, Luthuli N, functions associated with azole resistance in Candida glabrata Lowman W, Govender NP. 2016. Two echinocandin-resistant results in enhanced virulence in mice. Antimicrob Agents Candida glabrata FKS mutants from South Africa. Med Chemother. 55:1852–1860. doi:10.1128/AAC.01271-10. Mycol Case Rep. 11:24–26. doi:10.1016/j.mmcr.2016.03.004. BIOFOULING 577

Pappas PG. 2006. Invasive candidiasis. Infect Dis Clin North Rodrigues CF, Henriques M. 2017. Liposomal and deoxycholate Am. 20:485–506. doi:10.1016/j.idc.2006.07.004. amphotericin B formulations: effectiveness against biofilm Pappas PG, Kauffman CA, Andes DR, Clancy CJ, Marr KA, infections of Candida spp. Pathogens. 6:13. Ostrosky-Zeichner L, Reboli AC, Schuster MG, Vazquez Rodrigues CF, Silva S, Henriques M. 2014. Candida glabrata: JA, Walsh TJ, et al. 2015. Clinical practice guideline for the a review of its features and resistance. Eur J Clin Microbiol management of candidiasis: 2016 update by the Infectious Infect Dis. 33:673–688. doi:10.1007/s10096-013-2009-3. Diseases Society of America. Clin Infect Dis. 62:e1–e50. Rodrigues CF, Rodrigues ME, Silva S, Henriques M. 2017. Pappas PG, Rotstein CMF, Betts RF, Nucci M, Talwar D, De Waele Candida glabrata biofilms: how far have we come? J Fungi. JJ, Vazquez JA, Dupont BF, Horn DL, Ostrosky-Zeichner L, 3:11. doi:10.3390/jof3010011. et al. 2007. Micafungin versus caspofungin for treatment of Scorzoni L, de Paula e Silva ACA, Marcos CM, Assato PA, de candidemia and other forms of invasive candidiasis. Clin Melo WCMA, de Oliveira HC, Costa-Orlandi CB, Mendes- Infect Dis. 45:883–893. doi:10.1086/520980. Giannini MJS, Fusco-Almeida AM. 2017. Antifungal Perlin DS. 2007. Resistance to echinocandin- class antifungal therapy: new advances in the understanding and treatment drugs. Drug Resist Updat. 10:121–130. doi:10.1016/j. of mycosis. Front Microbiol. 8:1–23. drup.2007.04.002. Scott LJ. 2012. Micafungin: a review of its use in the prophylaxis Perlin DS. 2014. Echinocandin resistance, susceptibility testing and treatment of invasive Candida infections. Drugs. and prophylaxis: implications for patient management. 72:2141–2165. doi:10.2165/11209970-000000000-00000. Drugs. 74:1573–1585. doi:10.1007/s40265-014-0286-5. Silva S, Rodrigues CF, Araújo D, Rodrigues M, Henriques Perlin DS. 2015a. Mechanisms of echinocandin antifungal M. 2017. Candida species biofilms’ antifungal resistance. J drug resistance. Ann N Y Acad Sci:1–11. doi:10.1111/ Fungi. 3:8. doi:10.3390/jof3010008. nyas.2015.1354.issue-1. Spreghini E, Orlando F, Sanguinetti M, Posteraro B, Giannini Perlin DS. 2015b. Echinocandin resistance in Candida. Clin D, Manso E, Barchiesi F. 2012. Comparative effects of Infect Dis. 61:S612–S617. doi:10.1093/cid/civ791. micafungin, caspofungin, and anidulafungin against a Perlin DS, Shor E, Zhao Y. 2015. Update on antifungal difficult-to-treat fungal opportunistic pathogen, Candida drug resistance. Curr Clin Microbiol Reports. 2:84–95. glabrata. Antimicrob Agents Chemother. 56:1215–1222. doi:10.1007/s40588-015-0015-1. doi:10.1128/AAC.05872-11. Perlin DS, Teppler H, Donowitz GR, Maertens JA, Baden Stevens DA, Ichinomiya M, Koshi Y, Horiuchi H. 2006. Escape LR, Milne S, Brown AJ, Gow NA. 2007. Resistance to of Candida from caspofungin inhibition at concentrations echinocandin-class antifungal drugs. Drug Resist Updat. above the MIC (paradoxical effect) accomplished by 10:121–130. doi:10.1016/j.drup.2007.04.002. increased cell wall chitin; evidence for beta-1,6-glucan Pfaller MA, Boyken L, Hollis RJ, Kroeger J, Messer SA, Tendolkar synthesis inhibition by caspofungin. Antimicrob Agents S, Diekema DJ. 2008. In Vitro susceptibility of invasive Chemother. 50:3160–3161. doi:10.1128/AAC.00563-06. isolates of Candida spp. to anidulafungin, caspofungin, and Taff H, Nett J, Andes D. 2012. Comparative analysis of Candida micafungin: six years of global surveillance. J Clin Microbiol. biofilm quantitation assays. Med Mycol. 50:214–218. doi:10. 46:150–156. doi:10.1128/JCM.01901-07. 3109/13693786.2011.580016. Pfaller MA, Boyken L, Hollis RJ, Kroeger J, Messer SA, Taylor PR, Tsoni SV, Willment JA, Dennehy KM, Rosas M, Tendolkar S, Jones RN, Turnidge J, Diekema DJ. 2010. Wild- Findon H, Haynes K, Steele C, Botto M, Gordon S, Brown type MIC distributions and epidemiological cutoff values GD. 2007. Dectin-1 is required for ?-glucan recognition for the echinocandins and Candida spp. J Clin Microbiol. and control of fungal infection. Nat Immunol. 8:31–38. 48:52–56. doi:10.1128/JCM.01590-09. doi:10.1038/ni1408. Pfaller MA, Castanheira M. 2016. Nosocomial candidiasis: Tortorano AM, Prigitano A, Lazzarini C, Passera M, Deiana antifungal stewardship and the importance of rapid ML, Cavinato S, De Luca C, Grancini A, Lo Cascio G, Ossi diagnosis. Med Mycol. 54:1–22. C, et al. 2013. A 1-year prospective survey of candidemia in Pham CD, Iqbal N, Bolden CB, Kuykendall RJ, Harrison LH, Italy and changing epidemiology over one decade. Infection. Farley MM, Schaffner W, Beldavs ZG, Chiller TM, Park BJ, 41:655–662. doi:10.1007/s15010-013-0455-6. et al. 2014. Role of FKS mutations in Candida glabrata: MIC Tragiannidis A, Dokos C, Lehrnbecher T, Groll AH. 2012. values, echinocandin resistance, and multidrug resistance. Antifungal chemoprophylaxis in children and adolescents Antimicrob Agents Chemother. 58:4690–4696. doi:10.1128/ with haematological malignancies and following allogeneic AAC.03255-14. haematopoietic stem cell transplantation. Drugs. 72:685– Pierce C, Vila T, Romo J, Montelongo-Jauregui D, Wall G, 704. doi:10.2165/11599810-000000000-00000. Ramasubramanian A, Lopez-Ribot J. 2017. TheCandida Uppuluri P, Chaturvedi A, Srinivasan G, Banerjee M, albicans biofilm matrix: composition, structure and Ramasubramaniam A, et al. 2010. Dispersion as an important function. J Fungi. 3:14. doi:10.3390/jof3010014. step in the Candida albicans biofilm developmental Ramage G, Rajendran R, Sherry L, Williams C. 2012. Fungal cycle. PLoS Pathog. 6:e1000828. doi:10.1371/journal. biofilm resistance. Int J Microbiol. 2012:528521. ppat.1000828. Rodrigues CF, Gonçalves B, Rodrigues ME, Silva S, Azeredo Viscoli C, Bassetti M, Castagnola E, Cesaro S, Menichetti F, J, Henriques M. 2017. The effectiveness of voriconazole in Ratto S, Tascini C, Giacobbe DR. 2014. Micafungin for the therapy of Candida glabrata’s biofilms oral infections and Its treatment of proven and suspected invasive candidiasis in Influence on the matrix composition and gene expression. children and adults: findings from a multicenter prospective Mycopathologia [Internet] [accessed 2017 Jun 21]; 182:653– observational study. BMC Infect Dis. 14:e24198. doi:10.1186/ 664. http://www.ncbi.nlm.nih.gov/pubmed/28439794 s12879-014-0725-7. 578 C. F. RODRIGUES ET AL.

Wiederhold NP. 2009. Paradoxical echinocandin activity: in vitro potency in the presence of sera. Antimicrob Agents a limited in vitro phenomenon? Med Mycol. 47(Suppl Chemother. 51:1616–1620. doi:10.1128/AAC.00105-07. 1):S369–S375. doi:10.1080/13693780802428542. Williams C, Ramage G. 2015. Fungal biofilms in human Wiederhold NP, Lewis RE. 2003. The echinocandin antifungals: disease. Adv Exp Med Biol. 831: 11–27. doi:10.1007/978-3- an overview of the pharmacology, spectrum and clinical 319-09782-4. efficacy. Expert Opin Investig Drugs. 12:1313–1333. Williams DW, Wilson MJ, Lewis MAO, Potts AJC. 1995. doi:10.1517/13543784.12.8.1313. Identification of Candida species by PCR and restriction Wiederhold NP, Najvar LK, Bocanegra R, Molina D, Olivo fragment length polymorphism analysis of intergenic M, Graybill JR. 2007. In vivo efficacy of anidulafungin and spacer regions of ribosomal DNA. J Clin Microbiol. caspofungin against Candida glabrata and association with 33:2476–2479.